Hugging Face's logo

qualcomm
/
Midas-V2

Depth Estimation
PyTorch
android
Model card Files
xet
 Community
qaihm-bot commited on
Commit
dde1dc9
·
verified ·
1 Parent(s): 75a1f90

v0.48.0

Browse files

See https://github.com/qualcomm/ai-hub-models/releases/v0.48.0 for changelog.

Files changed (1) hide show
  1. README.md +68 -68
README.md CHANGED
@@ -14,7 +14,7 @@ pipeline_tag: depth-estimation
14
  Midas is designed for estimating depth at each point in an image.
15
 
16
  This is based on the implementation of Midas-V2 found [here](https://github.com/isl-org/MiDaS).
17
- This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/midas) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).
18
 
19
  Qualcomm AI Hub Models uses [Qualcomm AI Hub Workbench](https://workbench.aihub.qualcomm.com) to compile, profile, and evaluate this model. [Sign up](https://myaccount.qualcomm.com/signup) to run these models on a hosted Qualcomm® device.
20
 
@@ -27,26 +27,26 @@ Below are pre-exported model assets ready for deployment.
27
 
28
  | Runtime | Precision | Chipset | SDK Versions | Download |
29
  |---|---|---|---|---|
30
- | ONNX | float | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/midas/releases/v0.47.0/midas-onnx-float.zip)
31
- | ONNX | w8a8 | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/midas/releases/v0.47.0/midas-onnx-w8a8.zip)
32
- | QNN_DLC | float | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/midas/releases/v0.47.0/midas-qnn_dlc-float.zip)
33
- | QNN_DLC | w8a8 | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/midas/releases/v0.47.0/midas-qnn_dlc-w8a8.zip)
34
- | TFLITE | float | Universal | QAIRT 2.43, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/midas/releases/v0.47.0/midas-tflite-float.zip)
35
- | TFLITE | w8a8 | Universal | QAIRT 2.43, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/midas/releases/v0.47.0/midas-tflite-w8a8.zip)
36
 
37
  For more device-specific assets and performance metrics, visit **[Midas-V2 on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/midas)**.
38
 
39
 
40
  ### Option 2: Export with Custom Configurations
41
 
42
- Use the [Qualcomm® AI Hub Models](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/midas) Python library to compile and export the model with your own:
43
  - Custom weights (e.g., fine-tuned checkpoints)
44
  - Custom input shapes
45
  - Target device and runtime configurations
46
 
47
  This option is ideal if you need to customize the model beyond the default configuration provided here.
48
 
49
- See our repository for [Midas-V2 on GitHub](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/midas) for usage instructions.
50
 
51
  ## Model Details
52
 
@@ -62,70 +62,70 @@ See our repository for [Midas-V2 on GitHub](https://github.com/quic/ai-hub-model
62
  ## Performance Summary
63
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
64
  |---|---|---|---|---|---|---
65
- | Midas-V2 | ONNX | float | Snapdragon® X Elite | 2.937 ms | 35 - 35 MB | NPU
66
- | Midas-V2 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 1.92 ms | 1 - 201 MB | NPU
67
- | Midas-V2 | ONNX | float | Qualcomm® QCS8550 (Proxy) | 2.736 ms | 0 - 7 MB | NPU
68
- | Midas-V2 | ONNX | float | Qualcomm® QCS9075 | 3.873 ms | 1 - 4 MB | NPU
69
- | Midas-V2 | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 1.484 ms | 0 - 160 MB | NPU
70
- | Midas-V2 | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.294 ms | 0 - 164 MB | NPU
71
- | Midas-V2 | ONNX | float | Snapdragon® X2 Elite | 1.314 ms | 35 - 35 MB | NPU
72
- | Midas-V2 | ONNX | w8a8 | Snapdragon® X Elite | 1.51 ms | 18 - 18 MB | NPU
73
- | Midas-V2 | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.926 ms | 0 - 94 MB | NPU
74
- | Midas-V2 | ONNX | w8a8 | Qualcomm® QCS8550 (Proxy) | 1.373 ms | 0 - 4 MB | NPU
75
- | Midas-V2 | ONNX | w8a8 | Qualcomm® QCS9075 | 1.506 ms | 0 - 3 MB | NPU
76
- | Midas-V2 | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.758 ms | 0 - 55 MB | NPU
 
77
  | Midas-V2 | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.648 ms | 0 - 58 MB | NPU
78
- | Midas-V2 | ONNX | w8a8 | Snapdragon® X2 Elite | 0.638 ms | 20 - 20 MB | NPU
79
- | Midas-V2 | QNN_DLC | float | Snapdragon® X Elite | 3.296 ms | 1 - 1 MB | NPU
80
- | Midas-V2 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 2.127 ms | 0 - 195 MB | NPU
81
- | Midas-V2 | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 12.117 ms | 1 - 159 MB | NPU
82
- | Midas-V2 | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 3.063 ms | 1 - 3 MB | NPU
83
- | Midas-V2 | QNN_DLC | float | Qualcomm® SA8775P | 4.248 ms | 1 - 161 MB | NPU
84
  | Midas-V2 | QNN_DLC | float | Qualcomm® QCS9075 | 4.492 ms | 1 - 3 MB | NPU
85
- | Midas-V2 | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 6.827 ms | 0 - 192 MB | NPU
86
- | Midas-V2 | QNN_DLC | float | Qualcomm® SA7255P | 12.117 ms | 1 - 159 MB | NPU
87
- | Midas-V2 | QNN_DLC | float | Qualcomm® SA8295P | 5.506 ms | 1 - 161 MB | NPU
88
- | Midas-V2 | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 1.616 ms | 1 - 161 MB | NPU
89
- | Midas-V2 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.274 ms | 1 - 166 MB | NPU
90
- | Midas-V2 | QNN_DLC | float | Snapdragon® X2 Elite | 1.589 ms | 1 - 1 MB | NPU
91
- | Midas-V2 | QNN_DLC | w8a8 | Snapdragon® X Elite | 1.481 ms | 0 - 0 MB | NPU
92
- | Midas-V2 | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.921 ms | 0 - 85 MB | NPU
93
- | Midas-V2 | QNN_DLC | w8a8 | Qualcomm® QCS6490 | 4.35 ms | 0 - 2 MB | NPU
94
- | Midas-V2 | QNN_DLC | w8a8 | Qualcomm® QCS8275 (Proxy) | 2.852 ms | 0 - 46 MB | NPU
95
- | Midas-V2 | QNN_DLC | w8a8 | Qualcomm® QCS8550 (Proxy) | 1.327 ms | 0 - 2 MB | NPU
96
- | Midas-V2 | QNN_DLC | w8a8 | Qualcomm® SA8775P | 1.599 ms | 0 - 50 MB | NPU
97
  | Midas-V2 | QNN_DLC | w8a8 | Qualcomm® QCS9075 | 1.46 ms | 2 - 4 MB | NPU
98
- | Midas-V2 | QNN_DLC | w8a8 | Qualcomm® QCM6690 | 9.001 ms | 0 - 173 MB | NPU
99
- | Midas-V2 | QNN_DLC | w8a8 | Qualcomm® QCS8450 (Proxy) | 1.855 ms | 0 - 81 MB | NPU
100
- | Midas-V2 | QNN_DLC | w8a8 | Qualcomm® SA7255P | 2.852 ms | 0 - 46 MB | NPU
101
- | Midas-V2 | QNN_DLC | w8a8 | Qualcomm® SA8295P | 2.278 ms | 0 - 45 MB | NPU
102
- | Midas-V2 | QNN_DLC | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.683 ms | 0 - 46 MB | NPU
103
- | Midas-V2 | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 1.591 ms | 0 - 172 MB | NPU
104
- | Midas-V2 | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.555 ms | 0 - 50 MB | NPU
105
- | Midas-V2 | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 0.73 ms | 0 - 0 MB | NPU
106
- | Midas-V2 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 2.124 ms | 0 - 232 MB | NPU
107
- | Midas-V2 | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 12.164 ms | 0 - 179 MB | NPU
108
- | Midas-V2 | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 3.048 ms | 0 - 2 MB | NPU
109
- | Midas-V2 | TFLITE | float | Qualcomm® SA8775P | 4.313 ms | 0 - 179 MB | NPU
110
  | Midas-V2 | TFLITE | float | Qualcomm® QCS9075 | 4.441 ms | 0 - 39 MB | NPU
111
- | Midas-V2 | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 6.815 ms | 0 - 218 MB | NPU
112
- | Midas-V2 | TFLITE | float | Qualcomm® SA7255P | 12.164 ms | 0 - 179 MB | NPU
113
- | Midas-V2 | TFLITE | float | Qualcomm® SA8295P | 5.52 ms | 0 - 159 MB | NPU
114
- | Midas-V2 | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 1.637 ms | 0 - 182 MB | NPU
115
  | Midas-V2 | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.273 ms | 0 - 179 MB | NPU
116
- | Midas-V2 | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.758 ms | 0 - 78 MB | NPU
117
- | Midas-V2 | TFLITE | w8a8 | Qualcomm® QCS6490 | 3.522 ms | 0 - 28 MB | NPU
118
- | Midas-V2 | TFLITE | w8a8 | Qualcomm® QCS8275 (Proxy) | 2.455 ms | 0 - 44 MB | NPU
119
- | Midas-V2 | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 1.058 ms | 0 - 12 MB | NPU
120
- | Midas-V2 | TFLITE | w8a8 | Qualcomm® SA8775P | 1.373 ms | 0 - 48 MB | NPU
121
- | Midas-V2 | TFLITE | w8a8 | Qualcomm® QCS9075 | 1.212 ms | 0 - 20 MB | NPU
122
- | Midas-V2 | TFLITE | w8a8 | Qualcomm® QCM6690 | 8.324 ms | 0 - 171 MB | NPU
123
- | Midas-V2 | TFLITE | w8a8 | Qualcomm® QCS8450 (Proxy) | 1.59 ms | 0 - 78 MB | NPU
124
- | Midas-V2 | TFLITE | w8a8 | Qualcomm® SA7255P | 2.455 ms | 0 - 44 MB | NPU
125
- | Midas-V2 | TFLITE | w8a8 | Qualcomm® SA8295P | 2.005 ms | 0 - 43 MB | NPU
126
- | Midas-V2 | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.596 ms | 0 - 50 MB | NPU
127
- | Midas-V2 | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 1.346 ms | 0 - 167 MB | NPU
128
- | Midas-V2 | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.489 ms | 0 - 49 MB | NPU
129
 
130
  ## License
131
  * The license for the original implementation of Midas-V2 can be found
 
14
  Midas is designed for estimating depth at each point in an image.
15
 
16
  This is based on the implementation of Midas-V2 found [here](https://github.com/isl-org/MiDaS).
17
+ This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/main/qai_hub_models/models/midas) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).
18
 
19
  Qualcomm AI Hub Models uses [Qualcomm AI Hub Workbench](https://workbench.aihub.qualcomm.com) to compile, profile, and evaluate this model. [Sign up](https://myaccount.qualcomm.com/signup) to run these models on a hosted Qualcomm® device.
20
 
 
27
 
28
  | Runtime | Precision | Chipset | SDK Versions | Download |
29
  |---|---|---|---|---|
30
+ | ONNX | float | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/midas/releases/v0.48.0/midas-onnx-float.zip)
31
+ | ONNX | w8a8 | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/midas/releases/v0.48.0/midas-onnx-w8a8.zip)
32
+ | QNN_DLC | float | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/midas/releases/v0.48.0/midas-qnn_dlc-float.zip)
33
+ | QNN_DLC | w8a8 | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/midas/releases/v0.48.0/midas-qnn_dlc-w8a8.zip)
34
+ | TFLITE | float | Universal | QAIRT 2.43, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/midas/releases/v0.48.0/midas-tflite-float.zip)
35
+ | TFLITE | w8a8 | Universal | QAIRT 2.43, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/midas/releases/v0.48.0/midas-tflite-w8a8.zip)
36
 
37
  For more device-specific assets and performance metrics, visit **[Midas-V2 on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/midas)**.
38
 
39
 
40
  ### Option 2: Export with Custom Configurations
41
 
42
+ Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/main/qai_hub_models/models/midas) Python library to compile and export the model with your own:
43
  - Custom weights (e.g., fine-tuned checkpoints)
44
  - Custom input shapes
45
  - Target device and runtime configurations
46
 
47
  This option is ideal if you need to customize the model beyond the default configuration provided here.
48
 
49
+ See our repository for [Midas-V2 on GitHub](https://github.com/qualcomm/ai-hub-models/blob/main/qai_hub_models/models/midas) for usage instructions.
50
 
51
  ## Model Details
52
 
 
62
  ## Performance Summary
63
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
64
  |---|---|---|---|---|---|---
65
+ | Midas-V2 | ONNX | float | Snapdragon® X2 Elite | 1.313 ms | 35 - 35 MB | NPU
66
+ | Midas-V2 | ONNX | float | Snapdragon® X Elite | 2.929 ms | 35 - 35 MB | NPU
67
+ | Midas-V2 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 1.923 ms | 0 - 200 MB | NPU
68
+ | Midas-V2 | ONNX | float | Qualcomm® QCS8550 (Proxy) | 2.737 ms | 1 - 4 MB | NPU
69
+ | Midas-V2 | ONNX | float | Qualcomm® QCS9075 | 3.882 ms | 1 - 4 MB | NPU
70
+ | Midas-V2 | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 1.483 ms | 0 - 164 MB | NPU
71
+ | Midas-V2 | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.293 ms | 1 - 165 MB | NPU
72
+ | Midas-V2 | ONNX | w8a8 | Snapdragon® X2 Elite | 0.631 ms | 20 - 20 MB | NPU
73
+ | Midas-V2 | ONNX | w8a8 | Snapdragon® X Elite | 1.49 ms | 18 - 18 MB | NPU
74
+ | Midas-V2 | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.93 ms | 0 - 95 MB | NPU
75
+ | Midas-V2 | ONNX | w8a8 | Qualcomm® QCS8550 (Proxy) | 1.334 ms | 0 - 23 MB | NPU
76
+ | Midas-V2 | ONNX | w8a8 | Qualcomm® QCS9075 | 1.522 ms | 0 - 3 MB | NPU
77
+ | Midas-V2 | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.777 ms | 0 - 59 MB | NPU
78
  | Midas-V2 | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.648 ms | 0 - 58 MB | NPU
79
+ | Midas-V2 | QNN_DLC | float | Snapdragon® X2 Elite | 1.583 ms | 1 - 1 MB | NPU
80
+ | Midas-V2 | QNN_DLC | float | Snapdragon® X Elite | 3.311 ms | 1 - 1 MB | NPU
81
+ | Midas-V2 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 2.139 ms | 0 - 196 MB | NPU
82
+ | Midas-V2 | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 12.13 ms | 1 - 159 MB | NPU
83
+ | Midas-V2 | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 3.054 ms | 1 - 2 MB | NPU
84
+ | Midas-V2 | QNN_DLC | float | Qualcomm® SA8775P | 4.242 ms | 0 - 160 MB | NPU
85
  | Midas-V2 | QNN_DLC | float | Qualcomm® QCS9075 | 4.492 ms | 1 - 3 MB | NPU
86
+ | Midas-V2 | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 6.81 ms | 0 - 193 MB | NPU
87
+ | Midas-V2 | QNN_DLC | float | Qualcomm® SA7255P | 12.13 ms | 1 - 159 MB | NPU
88
+ | Midas-V2 | QNN_DLC | float | Qualcomm® SA8295P | 5.487 ms | 1 - 156 MB | NPU
89
+ | Midas-V2 | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 1.613 ms | 0 - 161 MB | NPU
90
+ | Midas-V2 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.271 ms | 1 - 165 MB | NPU
91
+ | Midas-V2 | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 0.71 ms | 0 - 0 MB | NPU
92
+ | Midas-V2 | QNN_DLC | w8a8 | Snapdragon® X Elite | 1.475 ms | 0 - 0 MB | NPU
93
+ | Midas-V2 | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.922 ms | 0 - 85 MB | NPU
94
+ | Midas-V2 | QNN_DLC | w8a8 | Qualcomm® QCS6490 | 4.189 ms | 0 - 2 MB | NPU
95
+ | Midas-V2 | QNN_DLC | w8a8 | Qualcomm® QCS8275 (Proxy) | 2.883 ms | 0 - 46 MB | NPU
96
+ | Midas-V2 | QNN_DLC | w8a8 | Qualcomm® QCS8550 (Proxy) | 1.309 ms | 0 - 2 MB | NPU
97
+ | Midas-V2 | QNN_DLC | w8a8 | Qualcomm® SA8775P | 1.603 ms | 0 - 50 MB | NPU
98
  | Midas-V2 | QNN_DLC | w8a8 | Qualcomm® QCS9075 | 1.46 ms | 2 - 4 MB | NPU
99
+ | Midas-V2 | QNN_DLC | w8a8 | Qualcomm® QCM6690 | 8.983 ms | 0 - 172 MB | NPU
100
+ | Midas-V2 | QNN_DLC | w8a8 | Qualcomm® QCS8450 (Proxy) | 1.908 ms | 0 - 80 MB | NPU
101
+ | Midas-V2 | QNN_DLC | w8a8 | Qualcomm® SA7255P | 2.883 ms | 0 - 46 MB | NPU
102
+ | Midas-V2 | QNN_DLC | w8a8 | Qualcomm® SA8295P | 2.245 ms | 0 - 45 MB | NPU
103
+ | Midas-V2 | QNN_DLC | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.677 ms | 0 - 46 MB | NPU
104
+ | Midas-V2 | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 1.596 ms | 0 - 169 MB | NPU
105
+ | Midas-V2 | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.565 ms | 0 - 51 MB | NPU
106
+ | Midas-V2 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 2.121 ms | 0 - 232 MB | NPU
107
+ | Midas-V2 | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 12.173 ms | 0 - 179 MB | NPU
108
+ | Midas-V2 | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 3.036 ms | 0 - 3 MB | NPU
109
+ | Midas-V2 | TFLITE | float | Qualcomm® SA8775P | 4.3 ms | 0 - 180 MB | NPU
 
110
  | Midas-V2 | TFLITE | float | Qualcomm® QCS9075 | 4.441 ms | 0 - 39 MB | NPU
111
+ | Midas-V2 | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 6.873 ms | 0 - 217 MB | NPU
112
+ | Midas-V2 | TFLITE | float | Qualcomm® SA7255P | 12.173 ms | 0 - 179 MB | NPU
113
+ | Midas-V2 | TFLITE | float | Qualcomm® SA8295P | 5.503 ms | 0 - 160 MB | NPU
114
+ | Midas-V2 | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 1.618 ms | 0 - 177 MB | NPU
115
  | Midas-V2 | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.273 ms | 0 - 179 MB | NPU
116
+ | Midas-V2 | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.76 ms | 0 - 79 MB | NPU
117
+ | Midas-V2 | TFLITE | w8a8 | Qualcomm® QCS6490 | 3.687 ms | 0 - 28 MB | NPU
118
+ | Midas-V2 | TFLITE | w8a8 | Qualcomm® QCS8275 (Proxy) | 2.439 ms | 0 - 46 MB | NPU
119
+ | Midas-V2 | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 1.078 ms | 0 - 3 MB | NPU
120
+ | Midas-V2 | TFLITE | w8a8 | Qualcomm® SA8775P | 1.371 ms | 0 - 47 MB | NPU
121
+ | Midas-V2 | TFLITE | w8a8 | Qualcomm® QCS9075 | 1.19 ms | 0 - 20 MB | NPU
122
+ | Midas-V2 | TFLITE | w8a8 | Qualcomm® QCM6690 | 8.343 ms | 0 - 171 MB | NPU
123
+ | Midas-V2 | TFLITE | w8a8 | Qualcomm® QCS8450 (Proxy) | 1.569 ms | 0 - 78 MB | NPU
124
+ | Midas-V2 | TFLITE | w8a8 | Qualcomm® SA7255P | 2.439 ms | 0 - 46 MB | NPU
125
+ | Midas-V2 | TFLITE | w8a8 | Qualcomm® SA8295P | 1.992 ms | 0 - 43 MB | NPU
126
+ | Midas-V2 | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.591 ms | 0 - 49 MB | NPU
127
+ | Midas-V2 | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 1.356 ms | 0 - 170 MB | NPU
128
+ | Midas-V2 | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.483 ms | 0 - 49 MB | NPU
129
 
130
  ## License
131
  * The license for the original implementation of Midas-V2 can be found